Fast and non-invasive serum detection technology based on surface-enhanced Raman spectroscopy and multivariate statistical analysis for liver disease.

This study explored a rapid and nondestructive liver disease detection technique based on surface-enhanced Raman spectroscopy (SERS) to realize the early diagnosis, prevention, and treatment of liver disease. SERS signals of serum were obtained from 304 normal individuals, 333 patients with hepatopathy, and 99 patients with esophageal cancer. The Raman spectra of different diseases were compared and diagnostic models of liver disease were established using orthogonal partial least squares discriminant analysis (OPLS-DA). The classification efficiencies of the different models were comprehensively evaluated through the receiver operating characteristic (ROC) curve and ten-fold cross validation. Area under the ROC curve is of greater than 0.97, indicating excellent classification of the groups. The accuracy rate of the test set reached 95.33%, and the lowest was 81.76% using the ten-fold cross validation. Thus, OPLS-DA combined with serum SERS is a rapid and non-invasive technique for the diagnosis of liver disease.